Dynamic sign language recognition based on convolutional neural networks and texture maps

Edwin Escobedo, Lourdes Ramirez, Guillermo Camara

Research output: Chapter in Book/Report/Conference proceedingPaper (Conference contribution)peer-review

12 Scopus citations

Abstract

Sign language recognition (SLR) is a very challenging task due to the complexity of learning or developing descriptors to represent its primary parameters (location, movement, and hand configuration). In this paper, we propose a robust deep learning based method for sign language recognition. Our approach represents multimodal information (RGB-D) through texture maps to describe the hand location and movement. Moreover, we introduce an intuitive method to extract a representative frame that describes the hand shape. Next, we use this information as inputs to two three-stream and two-stream CNN models to learn robust features capable of recognizing a dynamic sign. We conduct our experiments on two sign language datasets, and the comparison with state-of-the-art SLR methods reveal the superiority of our approach which optimally combines texture maps and hand shape for SLR tasks.

Translated title of the contributionReconocimiento dinámico de lenguaje de señas basado en redes neuronales convolucionales y mapas de textura
Original languageEnglish
Title of host publicationProceedings - 32nd Conference on Graphics, Patterns and Images, SIBGRAPI 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages265-272
Number of pages8
ISBN (Electronic)9781728152271
DOIs
StatePublished - Oct 2019
Externally publishedYes
Event32nd SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2019 - Rio de Janeiro, Brazil
Duration: 28 Oct 201931 Oct 2019

Publication series

NameProceedings - 32nd Conference on Graphics, Patterns and Images, SIBGRAPI 2019

Conference

Conference32nd SIBGRAPI Conference on Graphics, Patterns and Images, SIBGRAPI 2019
Country/TerritoryBrazil
CityRio de Janeiro
Period28/10/1931/10/19

Keywords

  • CNN
  • Sign language
  • Texture maps

Fingerprint

Dive into the research topics of 'Dynamic sign language recognition based on convolutional neural networks and texture maps'. Together they form a unique fingerprint.

Cite this